Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 484 views

    ANACOVA

    Uploaded by

    Putu Indra
    This document provides instructions for conducting an analysis of covariance (ANCOVA) using SPSS to statistically control for pre-test scores when comparing post-test scores between three treatment groups. The key assumption of homogeneity of regression slopes must be tested first to ensure the relationship between the covariate and dependent variable does not differ between groups. Conducting the ANCOVA with the covariate results in a better model fit and significant group differences, while omitting the covariate would not show significant differences between groups.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    ANACOVA

    Uploaded by

    Putu Indra
    484 views3 pages
    This document provides instructions for conducting an analysis of covariance (ANCOVA) using SPSS to statistically control for pre-test scores when comparing post-test scores between three treatment groups. The key assumption of homogeneity of regression slopes must be tested first to ensure the relationship between the covariate and dependent variable does not differ between groups. Conducting the ANCOVA with the covariate results in a better model fit and significant group differences, while omitting the covariate would not show significant differences between groups.

    Original Description:

    anacova analysis is useful for conducting experimental research actually in educational research.

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides instructions for conducting an analysis of covariance (ANCOVA) using SPSS to statistically control for pre-test scores when comparing post-test scores between three treatment groups. The key assumption of homogeneity of regression slopes must be tested first to ensure the relationship between the covariate and dependent variable does not differ between groups. Conducting the ANCOVA with the covariate results in a better model fit and significant group differences, while omitting the covariate would not show significant differences between groups.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    484 views3 pages

    ANACOVA

    Uploaded by

    Putu Indra
    This document provides instructions for conducting an analysis of covariance (ANCOVA) using SPSS to statistically control for pre-test scores when comparing post-test scores between three treatment groups. The key assumption of homogeneity of regression slopes must be tested first to ensure the relationship between the covariate and dependent variable does not differ between groups. Conducting the ANCOVA with the covariate results in a better model fit and significant group differences, while omitting the covariate would not show significant differences between groups.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    Analysis of Covariance - SPSS

    Analysis of Covariance Using SPSS

    Consider the following data. Column 1, variable a, represents three different treatment conditions. X is a pre-test score, and Y is a post-test score, a score that is obtained after treatment. We can use ANCOVA to statistically control for the pre-test scores. This is a relatively common use of ANCOVA when one wants to statistically equate all subjects on their pre-test scores.
    1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 10 6 5 8 9 4 9 12 4 8 8 8 6 11 10 9 7 8 7 3 6 8 6 8 15 1 4 6 10 0 7 13 6 13 5 18 9 7 15 15 14 9 16 7 13 18 13 6

    Using SPSS Syntax: GLM y BY a WITH x [requests a general linear model, y as dependent, a as independent, and x as covariate] /EMMEANS TABLES (a). [requests estimated marginal means, for factor a]

    Between-Subjects Factors N a 1.00 2.00 3.00 8 8 8

    Analysis of Covariance - SPSS

    Tests of Between-Subjects Effects Dependent Variable: y Source Corrected Model Intercept x a Error Total Corrected Total Type III Sum of Squares 275.828(a) .665 163.828 165.793 334.172 3010.000 df 3 1 1 2 20 24 Mean Square 91.943 .665 163.828 82.897 16.709 F 5.503 .040 9.805 4.961 Sig. .006 .844 .005 .018

    610.000 23 a R Squared = .452 (Adjusted R Squared = .370)

    Estimated Marginal Means


    a Dependent Variable: y 95% Confidence Interval a 1.00 2.00 3.00 Mean 6.531(a) 10.375(a) 13.094(a) Std. Error 1.453 1.459 1.487 Lower Bound 3.500 7.331 9.993 Upper Bound 9.562 13.418 16.196

    Questions 1. For the current analysis, what key assumption must be tested before interpreting the above output? Test for this assumption using the following syntax, GLM y BY a WITH x /DESIGN = a x a*x . and comment on whether or not it is satisfied. Notice that in the above syntax, we are specifying a test for factor a, x and the a*x interaction term. (hint: the outcome of the interaction test should tell you something about the assumption of homogeneity of regressions).

    2. Obtain the same output as above, including the test for homogeneity of regressions, using window-based commands.

    3. Draw appropriate conclusions from the present analysis, and comment on the overall goal of analysis of covariance.

    4. Had we not included the covariate X, would the results have changed? If so, how? Had we performed the ANOVA without the inclusion of covariate X, we would have obtained:

    Analysis of Covariance - SPSS

    Tests of Between-Subjects Effects Dependent Variable: Y Source Corrected Model Intercept a Error Total Corrected Total Type III Sum of Squares 112.000(a) 2400.000 112.000 498.000 3010.000 610.000 df 2 1 2 21 24 23 Mean Square 56.000 2400.000 56.000 23.714 F 2.361 101.205 2.361 Sig. .119 .000 .119

    a R Squared = .184 (Adjusted R Squared = .106)

    Comment on how the above output is different from what you obtained when the covariate X was included in the model.
    End of assignment.

    You might also like